Infinite Series

Discipline Course - I

Mathematics

Paper: Analysis - I

Lesson: Infinite Series

Course Developer: Chaitanya Kumar

Department/ College : Department of Mathematics, Delhi College of

Arts and Commerce (D.U.)

Institute of Lifelong Learning, University of Delhi pg. 1

 Infinite Series

Table of Contents:
Chapter : Infinite Series
 1: Learning Outcomes
 2: Introduction
 3: Infinite Series
o 3.1: Sequence of Partial sums of an infinite series
o 3.2. Convergence and Divergence of an infinite series
o 3.3. Comparison Series (p - series)
 4. Cauchy's General Principle of convergence for series
 5. Comparison Tests
o 5.1. First Comparison Test
o 5.2. Second Comparison Test

o 5.3. Comparison Test (Limit form)

 Summary
 References for further readings

1. Learning Outcomes

After you have read this chapter, you should be able to

 Define the infinite series,

 Sequence of partial sums of an infinite series,
 Convergence and Divergence of an infinite series,
 Comparison Series (p - series).
 Cauchy's General Principle of convergence for series.
 Comparison Tests

Institute of Lifelong Learning, University of Delhi pg. 2

 Infinite Series

2. Introduction :

The purpose of this chapter is to discuss sums that contain infinitely many
terms. Our first objective is to define what is meant by the "Sum" of
infinitely many real numbers. We begin with some terminology.

3. Infinite Series:

An expression of the form

u1 + u2 + u3 + ---------- + un+ ------

where each un is a real number, is called an infinite series of real numbers.


It is denoted by  un or  un , where un is called nth term of the series  un .
n 1

3.1. Sequence of Partial sums of an infinite series:

Let u1 + u2 + u3 + ---------- + un+ ------ be an infinite series. If Sn denotes

the sum of the first n terms of this series so that

Sn = u1 + u2 + u3 + ---------- + un

then <Sn> is called the sequence of partial sums of the given series.

3.2. Convergence and Divergence of an infinite series:

Definitions:

(i) An infinite series  un is said to be convergent, if the sequence

<Sn> of partial sums of  un is convergent. If lim Sn  S , then
n

S is called the sum of the series  un .

(ii) An infinite series  un is said to be divergent, if the sequence

<Sn> of partial sums of  un is divergent.

(iii) An infinite series  un is said to oscillate, if the sequence <Sn>

of partial sums of  un oscillates.

2 n 1
1 1 1
Example 1: Show that series 1          
3  3  3
converges.

Institute of Lifelong Learning, University of Delhi pg. 3

 Infinite Series

2 n 1
1 1 1
Solution : Here, Sn  1         
3  3  3
n n
 1  1
1   1 
  3   3
1 1 2
3 3
n
3 31
   
2 23

 1 
n

 lim Sn  lim  3  3     3
n n  2 2 3  2
 

Since the sequence <Sn> converges to 3 , therefore the

2
given series is convergent.

Example 2: Prove that series 1  2  3       n   diverges.

Solution : Here,

Sn  1  2  3      n
n  n  1

2

n  n  1
 lim sn  lim 
n n 2
Hence <Sn> diverges and consequently the given series
diverges.

Example 3: Show that series 2 - 2 + 2 + ---- oscillates

Solution : Here,

0 , if n is even
Sn  
2, if n is odd

Therefore, the sequence <Sn> oscillates and consequently

the given series oscillates.

Institute of Lifelong Learning, University of Delhi pg. 4

 Infinite Series

Example 4 : Show that series 1 + r + r2 + ------ (r > 0) converges if

0 < r < 1 and diverges if r  1.

Solution : Since r > 0, therefore, the series of positive terms

(i) if 0  r  1 , then

Sn  1  r  r 2      r n  1
1  rn 1 rn
  
1 r 1 r 1 r

 lim Sn 
n
1

1

lim r n
1 r 1 r n 
1 
 lim r n  0, if r 1 
1  r  n   

Thus <Sn> is a convergent sequence consequently

the given series is convergent for 0<r<1

(ii) When r = 1, then

Sn = 1 + 1 + 1 + ----- + 1 = n
 lim Sn  lim n  
n n

Thus sequence <Sn> is divergent consequently the

given series is divergent for r = 1.

(ii) When r > 1, then

Sn = 1 + r + r2 + - - - + rn-1 > n
 lim Sn  
n

Thus <Sn> is divergent consequently the given

series is divergent when r > 1.

Value Addition: Remark

An infinite series of the form a + a.r + a.r2 + ------, is called a geometric
series with common ration r.
As an application of Example 4, we see that
1 1 1 1  1 
(a)  n   2  3     is convergent  since r   1
3 3 3 3  3 
(b) The series 1  1  1     is divergent  since r  1
(c)  4n  4  42  43     is divergent  since r  4  1

Institute of Lifelong Learning, University of Delhi pg. 5

 Infinite Series


Example 5 : Show that series   1 n is not convergent.
n

n 1

Solution : We have

  1 . n   1  2  3  4  5  6    
n

n 1

 S1   1, S2   1  2  1, S3   1  2  3   2
S4   1  2  3  4  2, S5   3     etc

  Sn   1, 1,  2, 2  3,       which is not bounded

Hence the sequence <Sn> is not convergent consequently

the given series is not convergent.

1 1 1
Example 6 : Show that series        is convergent
1.2 2.3 3.4

Solution : We have

1 1 1 1
Sn      
1.2 2.3 3.4 n  n  1

 1 1 1 1 1 1 1 
 1                   
 2  2 3 3 4 n n  1
1
 1
n 1
 1  1
 lim Sn  lim 1   1 1
n n
 n  1 
Therefore, the given series converges to 1.

3.3: Comparison Series (p - series)

1 1 1 1 1
Theorem 1 : The series  p
 p  p  p      p     is convergent
n 1 2 3 n
if p>1 and divergent if p  1 .

Proof : Let Sn denote the sum of the first n terms of the given series.

Case I, when p>1

Institute of Lifelong Learning, University of Delhi pg. 6

 Infinite Series

Now,

1
 1
1p
1 1 2 1
p
 p  p  p 1
2 3 2 2
----------------------------
----------------------------

2 n 1  2 n
n
1 1 1 1  1 
        p 1 
2  2  1 2  1 2  2 
p 1
2 
n p n p n 1 p n p n

Adding,
2 n
1  1   1 
S 2n  1  1  1    p 1         p 1 
2 p 1 2  2 
n 1   1  
n 1
 1  p 1
1   p 1  
1   p 1  2
2    2   2 p 1
   p 1 ,  n
1
1  p 1
1
1 2  1
2 2 p 1

we know that when n is any positive integer

2n 1 1  2n  n

2 p 1
 Sn  S2n  S2n 1 1 
2 p 1 1

2 p 1
Since for a given p, is a fixed number.
2 p 1  1

Hence the sequence <Sn> is bounded above and hence the given series
converges for p>1

Case II when p  1

We know, if n is any positive integer and p  1, then

1
np  n   1
np n

Institute of Lifelong Learning, University of Delhi pg. 7

 Infinite Series

1 1
 1 1  1
2 p
2 2

1 1 1 1 2 1
 p
 p    
3 4 3 4 4 2

1 1 1 1 1 1 1 1 4 1
 p
 p  p  p      
5 6 7 8 5 6 7 8 8 2
-----------------------------------------------
-----------------------------------------------

1 1 1 1 1 1
       
2 m 1
 1 2 m 1
 2
p
2 
m p
2 m 1
1 2 m 1
2 2m

2m  2m1 1
 
2m 2

Adding ,

m
S 2m 
2

We shall show that <Sn> is not bounded above if G be any number, however
large then there exist mN such that

m
G
2
Let n  2m
 Sn  S2m  G

Thus the sequence <Sn> of partial sums of the given series is not bounded
above and hence the series diverges for p  1.

1
Thus the given series  converges if p > 1 and diverges if p  1.
np

Value addition : Note

The p-series is extensively used to examine the convergence and divergence
of the large number of series. The students are strongly advised to
remember the conditions for the convergence and divergence of the p-
series.

Institute of Lifelong Learning, University of Delhi pg. 8

 Infinite Series

Example 7:

1 1 1 1
1.  2
 2  2  2      converges  p  2  1
n 1 2 3

1 1 1
2.   1           diverges  p  1
n 2 3

3. 
1
n
3
2
converges  p  3 1
2 
1 1 1  1 
4.  1        diverges  p  1
n 2 3  2 

3.4. A necessary condition for convergence:

Theorem 2 : A necessary condition for a series  un to converge is that

un  0 as n  .

Proof : Let Sn  u1  u2      un

then un  Sn  Sn 1 _________________ (i)

If Sn  S as n  , then taking limits of both sides of (i), we obtain

lim un  lim  Sn  Sn 1 
n  n 

 lim Sn  lim Sn 1
n  n 

 S S  0

Value Addition: Remark

The students should note that the above condition is only necessary, not
sufficient. There exist non-convergent series for which un  0 as n  , for
example, consider the series
1 1 1 1
  1       
n 2 3 n
1 1
Hence un  which tends to zero as n   but the series  is divergent.
n n
Corollary : If lim un  0, then the series  un cannot converge.
n 

Institute of Lifelong Learning, University of Delhi pg. 9

 Infinite Series

n 1 2 3
Example 8: Show that the series         is not convergent.
n 1 2 3 4

Solution : We have

n 1
un  
n 1 1 1
n

1 1
 lim un  lim  1 0
n  n 
1
1 1 0
n

Hence the given series is not convergent by above corollary.

Example 9 : Show that the series

1 2 n
     
4 6 2  n  1
does not converge.

Solution : Here,

 n  1 n
un    
 2  n  1  2 n 1
 

1 1

2 1 1
n

1 1 1
 lim un  lim  0
x x 2 1 1 2
n

Hence the given series does not converge.

1
Example 10 : Test for convergence the series  cos  
n

Solution : We have

Institute of Lifelong Learning, University of Delhi pg. 10

 Infinite Series

1
un  cos
n

1
 lim un  lim cos
n  n n

 cos 0  1  0

Hence the given series does not converge.

4. Cauchy's General Principle of convergence for series:

Theorem 3: A necessary and sufficient condition for a series  un to

converge is that to each >0, there exist a positive integer m, such that

um  1  um  2      un   , for all n  m.

Proof : The series  un converges iff the sequence <Sn> of its partial sums
converges.

By Cauchy's general participle of convergence for sequences, a necessary

and sufficient condition for the convergence of <Sn> is that given any >0
there exists a positive integer m such that.

sn  Sm   , whenever n  m

 um  1  um  2      un   , whenever n  m.

Hence the theorem.

Example 11 : Show the series

1 1 1 1
  1            does not converge.
n 2 3 n

Solution : Suppose that the given series converges. Then the sequence
<Sn> of partial sums of the given series is convergent. By Cauchy's principle
of convergence for sequences for   1 , there exist a positive integer m
2
such that

1
sn  Sm  , v n  m.
2

Institute of Lifelong Learning, University of Delhi pg. 11

 Infinite Series

 1 1 1 1 1 1
 1             
 2 3 m m 1 m  2 n

 1 1 1  1
 1           ,nm
 2 3 m 2

1 1 1 1
     , v n  m
m 1 m  2 n 2

On taking n = 2m, we have

1 1 1 1 1 1 1 1
     ...   m. 
m 1 m  2 2m 2m 2m 2m 2m 2

1 1 1 1
      n  2m  m 
m 1 m  2 n 2

Thus we have a contradiction. Hence the given series does not converge.

Example 12 : Show the series

1 1 1 1
  1            does not converge.
n 2 3 n

Solution : Suppose that given series is convergent. Therefore the

sequence <Sn> of partial sums of the given series is convergent.

1
 By Cauchy's general principle of convergence for any   , there exist a
2
positive integer m such that

1
sn  Sm  , v nm
2

 1 1 1 1 1
 1            
 2 m m 1 m  2 n

 1 1  1
 1          , v nm
 2 m 2

1 1 1 1
        , v n  m _____________ (1)
m 1 m  2 n 2

Institute of Lifelong Learning, University of Delhi pg. 12

 Infinite Series

on taking n = 2m, we have

1 1 1 1 1 1
     . . .
m 1 m  2 n m 1 m  2 2m

1 1 1 1 1
      m. 
2m 2m 2m 2m 2

Hence

1 1 1 1
     n  2 m  m  ______ (2)
m 1 m  2 n 2

Since (1) and (2) are contradictory statements, hence the given series does
not converge.

5. Comparison Tests:

5.1. First Comparison Test : Let  un and  vn be two positive terms

series such that

un  kvn , v n  m       1

Where k is fixed positive number and m is a fixed positive integer then

(i)  vn converges   un converges,

(ii)  un diverges   vn diverges

Proof : Let  Sn  and Tn  be the sequences of partial sums of the series
 un and  vn respectively for n  m, we have

Sn  Sm   u1  u2      um  um 1  um 2      un    u1  u2      um 

 Sn  Sm  um 1  um 2      un   2
Similarly, Tn  Tm  vm 1  vm 2      vn   3
From equation (1) and (2), we have

Sn  Sm  k  vm 1  vm  2      vn 

 Sn  Sm  k Tn  Tm 

Institute of Lifelong Learning, University of Delhi pg. 13

 Infinite Series

 Sn  Sm  kTn  a         4

where a  Sm  k Tm is a fixed number

(i) Suppose  vn converges

By theorem, the sequence Tn  of partial sums of  vn is bounded above i.e.

there exists a positive real number t such that

Tn  t , v n         5

From equations (4) and (5), we have

Sn  k t ,  a , v n

Thus the sequence  Sn  of partial sums of  un is bounded above. Hence  un

is convergent.

(ii) Suppose  un diverges, then

lim Sn          6
x 

From equations (4) and (6), we have

1
Tn   sn  a 
k

 lim Tn   
x 
 k  0

 The sequence Tn  of partial sums of the series  vn diverges.

Hence  vn is divergent

Example 13 : Show the series  e n is convergent.

2

Solution : We know that

1 1
n2

e n2

Institute of Lifelong Learning, University of Delhi pg. 14

 Infinite Series

1
 e n 
2
, vn
n2

1
But  is convergent since p = 2 > 1
n2

Hence by first comparison Test  e n is also convergent.

2

 1
Example 14 : Show the series  is divergent.
n2 log n

Solution : We know that log n  n, v n  2

1 1
  , vn2
log n n

1 1
  , vn2
n log n

1
But  is divergent since p=1 consequently by First comparison Test
n
 1
 is also divergent.
n2 log n

Example 15 : Test for convergence the series

1 1 1 1
1 2
 3  4    n  
2 3 4 n

Solution : We know that

n n  2n , v n  2

1 1
 n
 n , v n 2
n 2

1 1 1 1
Since  n
  2  3   is a geometric series with common ratio
2 2 2 2
1
r  1
2

Institute of Lifelong Learning, University of Delhi pg. 15

 Infinite Series

1 1
Since  n
is convergent. Therefore by first comparison test,  n is also
2 n
convergent.

5.2. Second Comparison Test: Let  un and  vn are two positive term
un v
series such that  n , v n  m      1
u n 1 v n 1

then (i)  vn converges   un converges

(ii)  un diverges   vn diverges

Proof : For n  m, we have

um u um 1 u m  2 un 1
 m . . 
un um 1 um  2 um  3 un

vm vm 1 vm  2 vn 1
 . .  , by 1
vm 1 vm  2 vm  3 vn

vm

vn

um vm un v
Hence  or  n , vnm
un vn um vm

 un  k vn , v n  m

um
where k  is a fixed positive number ------- (2)
vm

Applying First comparison Test in (2), we obtain

(i)  vn converges   un converges,

(ii)  un diverges   vn diverges

5.3. Comparison Test (Limit form):

Institute of Lifelong Learning, University of Delhi pg. 16

 Infinite Series

un
Let  un and  vn be two positive term series such that lim  , (ℓ is
x  v
n

non zero finite number) then  un and  vn both converge or diverge

together.

un
Proof : Since  0, v n
vn

un
 lim 0
x  vn

Thus  0, as 0

Let  > 0 be any number such that ℓ   > 0

un
since lim  , so there exist a positive integer m such that
x  v
n

un
 , vn  m
vn

un
      , v n  m, where vn  0, v n
vn

     vn  un      vn v n  m      1

From equation (1), we consider

un      vn v n  m       2

Where ℓ +  is a fixed positive number

Applying first comparison test in (2), we obtain

 vn converges   un converges 
     A
 un diverges   vn diverges 

From equation (1), we consider

    vn  un vn m

vn  k un vn m ............. (3)

Institute of Lifelong Learning, University of Delhi pg. 17

 Infinite Series

1
where k   0 is a fixed number.
( )

Applying first comparison test in (3), we get

 un converges   vn converges 
     B
 vn diverges   un diverges 

From (A) and (B), we have

 un converges   vn converges 

 un diverges   vn diverges 

Hence, the two series u n and v

n converge or diverge together.

Value Addition
If  0or   , then the conclusion of the above test may not hold good.
1 1 u n 1
(1) Let  un   2 and  vn   , then lim n  lim 2  lim  0 , Thus  0 ,
n n n  v n  n n  n
n

but in this case u n converges and v n diverges.

1 1 un n2
(II) Let  un   n
and  vn   n2
, then lim
n  v
 lim
n  n
 lim n   , Thus
n 
n

  , but in this case u n diverges and v n converges.

1 2 3 4
Example 16: Test for convergence the series    . . .
5 7 9 11

n
Solution: Here, un   0, n
2n  3

n 1
for large values of n, un behaves as i.e. as .
2n 2 n

1
Let vn 
n

un n n n 1
 lim  lim  lim 
n  v n  (2n  3) 1 n  (2n  3) 2
n

Institute of Lifelong Learning, University of Delhi pg. 18

 Infinite Series

u v
un
since lim is non-zero finite number. Therefore, the series n and n
n  v
n

1
converge and diverse together. But the series  n
diverges. Hence the

given series also diverges.

1 1 1 1
Example 17: Test for convergence the series    . . .
3.7 4.9 5.11 6.13

Solution: We have

1 1
un  
(n  2)(2n  5)  2  5
n 2 1   2  
 n  n

1
Let vn  , then
n2

u n2 1
 lim n  lim 
n  v n   2  5 2
n
n 2 1   2  
 n  n

u
un
since lim  0 is a non-zero finite number, therefore, the series n and
n  v
n

v
1
n converge and diverse together. But the series v   n
n 2
converges.

Hence the given series also converges.

Example 18: Test the convergence of the series whose nth term is
n4  1  n2 .

Solution: Here, un  n4  1  n2  0, n

therefore, the given series is of positive terms

Institute of Lifelong Learning, University of Delhi pg. 19

 Infinite Series

un  n 4  1  n 2
1
 1  2
 n 1  4   n 2
2

 n 
Also,
 1 
 n 2 1  4  . . .   n 2
 2n 
1
 2 . . .
2n

1
for sufficiently large values of n, un behaves as . Therefore,
2 n2

1
Let vn 
n2

un 1
 lim 
n  vn 2

un 1
since lim  , which is non-zero finite number, therefore, by comparison
n  v 2
n

test series u n and v n converge or diverse together. But the series v n

converges. Hence the given series also converges.

1 1
Example 19: Show that the series  n
tan is convergent.
n

1 1
Solution: Here, un  tan
n n

1
Let vn 
n n

1
tan
u n 1  0
 lim n  lim
n  v
n
n  1
n

un
since lim  1 , which is non-zero finite number, therefore, by comparison
n  v
n

test series u n and v n converge or diverse together. But the series v n

converges. Hence the given series also converges.

Institute of Lifelong Learning, University of Delhi pg. 20

 Infinite Series

1 3 5
Example 20: Test for convergence the series   . . .
1.2.3 2.3.4 3.4.5

Solution: We have

2n  1
un 
n(n  1)(n  2)

2
for large values on n, un 
n2

1
Let vn  , then
n2

un (2n  1)n 2  n  2n  1 
   
vn n(n  1)(n  2)  n  1  n  2 
  1
 1  2  n 
   2
1 2
 1   1  
 n  n

un
since lim  2  0 is a non-zero finite number, therefore, by the comparison
n  v
n

test (Limit form), the series u n and v

n converge or diverse together.
1
Since the series v   n
n 2
converges. Hence the given series also

converges.

Exercise:

Test each of the series 1- 10 for convergence

1 1 1
1.   . . .
1.2 2.3 3.4

1 1 1
2.   . . .
1.3.5 2.4.6 3.5.7

1  1 1
3.  2n  1
n 1
 Hint un  2n  1 , vn  n 

Institute of Lifelong Learning, University of Delhi pg. 21

 Infinite Series


n2  1
4. 
n 1 n  4
3  Hint: vn  n 

1 1 1
5.   . . .
2 3 3 4 4 5

1 1 1
6.    ...
1  2 1  2 1  23
2

 1 1 
7. n  
 n  1 
2 2
n 1 

1 3 5 7
8.     ...
4.6 6.8 8.10 10.12

1 1 1 1
9.     ...
1.2 2.3 3.4 4.52
2 2 2


1
10. 2 n 1
n
x
for all positive values of x.

11. Test for convergence the series whose nth term is

(a) n 1  n

(b) n3  1  n3

(c) n4  1  n4  1

(d) n2  1  n

1
12. Prove that the series n
n2
2
log n
converges.

Summary:

In this lesson we have emphasize on the followings:

 Define the infinite series,

 Sequence of partial sums of an infinite series,

Institute of Lifelong Learning, University of Delhi pg. 22

 Infinite Series

 Convergence and Divergence of an infinite series,

 Comparison Series (p - series).
 Cauchy's General Principle of convergence for series.
 Comparison Tests
References:

1. Bartle R.G. and Sherbert D.R., Introduction to Real Analysis (3rd

edition), John Wiley and Sons (Asia) Pvt. Ltd. Singapore, 2002.

2. Brian S. Thomson, Andrew, M. Bruckner and Judith B. Bruckner,

Elementary Real Analysis, Prentice Hall, 2001.

3. Gerald G. Bilodeau, Paul R. Thie, G.E. Keough, An Introduction

to Analysis, Joness Bartlett. Second Edition, 2010.

Institute of Lifelong Learning, University of Delhi pg. 23